Animal & Veterinary
The Pork Industry by Steve Larsen, Ph.D., National Pork Board
DR. LARSEN: First, I would like to thank the organizing committee to allow the pork industry to come and give our perspective on a NARMS survey. So I guess here is the point in the program that we are going to stop talking about virulence genes and spa’s and ma’s and jah’s and things like that and get into what the industry sees in a sampling scheme.
So I plan to present today is kind of what the pork industry envisions a NARMS sampling scheme or sampling program, maybe discuss some sampling points or current sampling points or future sampling points that we have some input on.
Another part is partnering, how we could partner in this whole scheme and make it successful.
Some recommendations and then, I don’t know if we have any time for questions, or maybe questions at the end of the session would be fine.
What the pork industry did was we got together what we call a pork safety committee that this kind of fits under. So we got together and talked about the NARMS. As Paula said yesterday, we have a good collaboration with Paula and her group. She came and talked to us about this so we were somewhat prepared to come up with some comments for today.
What the pork safety committee -- what the pork industry put together is first we would really want to get the objectives defined. Define the objectives first. Just like we are doing today. We have had other NARMS meetings in the past and it has not been very publicized of what input was given to the objectives and what isn’t. So define the objectives and how they were shaped, with stakeholder input, just like what we are doing today.
Then state the sampling plan. So once the objectives are defined clearly with stakeholder input, state the sampling plan so we can see what it is like because sometimes it has not been that clear. You know, we have the D-Lab samples, the slaughter house samples, and now we are talking about on-farm samples.
But we want to make sure that sampling plan answers those objectives. If you look now, I believe there are four objectives. Will they stay the same? Will they change? How will they change if they do change? Make sure it is statistically valid. Then we want to understand why samples, how many samples and why, just to make sure that everything goes back to those objectives and we are meeting those objectives that everybody gave input on.
We would also like to know before we start sampling how those results will be analyzed and presented. Because as you see today, there are so many data points out there, so many good analyses that we need to get that out so we can get everybody’s input to make sure that we are still going along that right path and make sure we get everybody’s input on it.
I guess the bottom line is to just make it as open as possible. You know, because we will talk about on-farm samples and one thing producers really like to see is openness. They do not want to see anything that could be construed as, well we will do this and then let you know after we do it, but we want to make it as open a process as possible.
For the sampling points, the first thing that we would like to say is we definitely want to make it a voluntary participation because when you talk about mandatory participation producers, at least if we go to the farm, they do not like to hear that word mandatory. They will shy away from any type of sampling scheme.
So there should be some incentives. Incentives could be anything. I don’t know if it would be monetary or access of the data or something like that to increase voluntary participation. So there should be some type of incentive. What incentives, we don’t know, but something to increase voluntary participation.
As we discussed yesterday, it appears that there may be three types of data streams into the NARMS program, farms, plants and diagnostic lab. We would really like to see those not only combined, but they need to be differentiated especially in the results because if you lump all these three together, it is like comparing apples with oranges to bananas. You really can’t do it and if you make a decision on a group data stream then you are probably going to have a biased result.
So I guess the bottom line of this is we want to decrease as much bias as possible. Each one of those is important and tells a very good story, but let’s differentiate it now because I don’t believe right now they are differentiated, but we need as many data points as we can.
We would also like to see regional identification. Within that regional identification, we need multiple data points per region. The fewer the data points, for example, if you have -- if we do go to farm or if we keep the Plant&D* lab, if you have one data point per region, than that increases your bias and you may see some huge data swings that is only explained by one data point.
So if you can get multiple data points per region, that is highly encouragable, especially if you have these three data streams. You are going to have multiple data points per data stream to avoid these regional biases and huge potential data swings of the data.
Partnering with NARMS. Of course, the Pork Board would need and love to be involved in partnering with NARMS specifically for producer education, because if there is going to be on-farm samples, or if this data is going to be used and there are no on-farm samples and there are just FSIS data or FSIS samples, because there is some sort of a bias towards those.
If you look at FSIS risk-based samples and those samples are taken in the cooler for carpet swabs, they go all the way through the process and you talk about antimicrobial resistance, you don’t know if that came from the farm or if that came from some sort of process in the plant. It can’t be defined.
So there is probably going to have to be some type of farm sample and it would be great to get Pork Board involved so we can do some producer education and talk to the producers to show the benefits of on-farm samples.
Along with that, we highly encourage the involvement of the American Associate of Swine Veterinarians because as every producer will come to the Pork Board for input, they will go, just like everybody else should, they will go seek a second opinion. That second opinion is the herd veterinarian.
So they will say, “Hey, the Pork Board told us about this great sampling scheme for this plan of NARMS. As my herd veterinarian, what do you think of that type of scheme?” SO it would be great to get AASV involved so the herd veterinarian doesn’t say, “Well I have never heard of that before” or “Here is what I think of it.” But we need to work together with the herd veterinarians and the National Pork Board.
Quickly some other recommendations that the pork industry came up with include it would be great if we could shorten the length of published results. I don’t know if that is every quarter, every six months, but right now if you look on the NARMS database, I believe their executive summary is back at 2007.
If one of the objectives is to give input or give decision makers or policy makers a way to decide on potential antibiotics, we are talking two- to three-year-old data. Is that really what we are aiming for? Is that really what we are shooting for?
If we do change something on the farm or something in the plant or something, that means the first time we see the data is three years after we have implemented something. So that means we need more than one data stream. So that means you are going to have to go to year two, so that is four years after we implemented something. Is that going to be enough to make a good decision? So we need to try to shorten the length of published results.
Then if you look back at the peer-published results, I think if you look on the website again, I think the last published article in a peer review was 2002 and it may be sooner than that, I don’t know. But on the NARMS website it says 2002. That is eight years old. So if we can just get that quicker and quicker turn-around that would be really great.
Then if we could have reviewed and analyzed data and then communicate cooperatively with ARS, CDC, FDA and the industry, so all the partners, so everybody sees what is coming out. Maybe it is a yearly meeting to talk about, “Okay, here is what we found in 2009. Here is what we are getting ready to publish and put on the internet.” Let’s get a communication put out there of what this means, how it means, and things like that.
Paula brought up some good results, talk about DT-104 and since the number of samples have decreased over time, but it appears that DT-104 has potentially increased, we don’t know if that is a factor of the same number of isolates are DT-104 but your overall sample is decreased so that is showing an increase, but really is not an increase. Or if they are both decreasing but the bigger number is decreasing more than the other number.
So that could be an inherent bias to the problem. So we need to know both the numerator and denominator and that would be communicated through this industry, ARS, CDC, FDA, hopefully joint meeting.
And just like what we are doing today, maybe a periodic review of NARMS just to give stakeholders input to make sure that the NARMS program is going down the right road, the correct road. It is always good to get this input to make sure we are getting out what we are doing.
And then just overall consistency. And that is pretty much all I had. Paula, hopefully that is exactly what you are looking for.
DR. FEDORKA-CRAY: Nothing like being criticized. That is good.
DR. LARSEN: Oh, I wasn’t criticizing. That was just input on the program.
DR. FEDORKA-CRAY: Oh, I know. Yes it was fine. Thank you.
DR. LARSEN: Thank you Paula.
DR. FEDORKA-CRAY: We will have discussion after the break. These are certainly some things that we definitely have to go over over time and we appreciate all of the cooperation from the industry. When we have asked to come and talk to them, they have certainly been welcome and open and open to ideas about revamping the sampling scheme.